Overview#

This notebook gives a general overview of the features included in the dataset.

Hide imports
%load_ext autoreload
%autoreload 2
import os

import dimcat as dc
import pandas as pd
import plotly.express as px
from dimcat import filters, plotting
from IPython.display import display

import utils
RESULTS_PATH = os.path.abspath(os.path.join(utils.OUTPUT_FOLDER, "overview"))
os.makedirs(RESULTS_PATH, exist_ok=True)


def make_output_path(
    filename: str,
    extension=None,
    path=RESULTS_PATH,
) -> str:
    return utils.make_output_path(filename=filename, extension=extension, path=path)


def save_figure_as(
    fig, filename, formats=("png", "pdf"), directory=RESULTS_PATH, **kwargs
):
    if formats is not None:
        for fmt in formats:
            plotting.write_image(fig, filename, directory, format=fmt, **kwargs)
    else:
        plotting.write_image(fig, filename, directory, **kwargs)

Loading data

D = utils.get_dataset("couperin_concerts", corpus_release="v2.3")
package = D.inputs.get_package()
package_info = package._package.custom
git_tag = package_info.get("git_tag")
utils.print_heading("Data and software versions")
print("François Couperin – Concerts Royaux version v2.3")
print(f"Datapackage '{package.package_name}' @ {git_tag}")
print(f"dimcat version {dc.__version__}\n")
D
Data and software versions
--------------------------

François Couperin – Concerts Royaux version v2.3
Datapackage 'couperin_concerts' @ v2.3
dimcat version 3.4.0
Dataset
=======
{'inputs': {'basepath': None,
            'packages': {'couperin_concerts': ["'couperin_concerts.measures' (MuseScoreFacetName.MuseScoreMeasures)",
                                               "'couperin_concerts.notes' (MuseScoreFacetName.MuseScoreNotes)",
                                               "'couperin_concerts.expanded' (MuseScoreFacetName.MuseScoreHarmonies)",
                                               "'couperin_concerts.chords' (MuseScoreFacetName.MuseScoreChords)",
                                               "'couperin_concerts.metadata' (FeatureName.Metadata)"]}},
 'outputs': {'basepath': None, 'packages': {}},
 'pipeline': []}
filtered_D = filters.HasHarmonyLabelsFilter(keep_values=[True]).process(D)
all_metadata = filtered_D.get_metadata()
assert len(all_metadata) > 0, "No pieces selected for analysis."
all_metadata
TimeSig KeySig last_mc last_mn length_qb last_mc_unfolded last_mn_unfolded length_qb_unfolded volta_mcs all_notes_qb ... has_drumset ambitus path originalFormat staff_1_instrument staff_1_ambitus staff_2_instrument staff_2_ambitus staff_3_instrument staff_3_ambitus
corpus piece
couperin_concerts c01n01_prelude {1: '4/4'} {1: 1} 25 23 98.0 25 23 98.0 () 219.00 ... 0 36-84 (C2-C6) MS3/c01n01_prelude.mscx xml Instrument 1 59-84 (B3-C6) Instrument 1 36-57 (C2-A3) <NA> <NA>
c01n02_allemande {1: '4/4'} {1: 1} 20 18 72.0 40 36 144.0 () 168.50 ... 0 31-83 (G1-B5) MS3/c01n02_allemande.mscx xml Instrument 1 59-83 (B3-B5) Instrument 1 31-66 (G1-F#4) <NA> <NA>
c01n03_sarabande {1: '3/4'} {1: -2} 30 28 90.0 56 56 168.0 (8], [9]], [[29], [30) 237.75 ... 0 31-81 (G1-A5) MS3/c01n03_sarabande.mscx xml Instrument 1 58-81 (Bb3-A5) Instrument 1 31-67 (G1-G4) <NA> <NA>
c01n04_gavotte {1: '2/2'} {1: -2} 18 14 60.0 32 28 112.0 (5], [6]], [[17], [18) 135.00 ... 0 38-79 (D2-G5) MS3/c01n04_gavotte.mscx xml Instrument 1 58-79 (Bb3-G5) Instrument 1 38-57 (D2-A3) <NA> <NA>
c01n05_gigue {1: '6/8'} {1: 1} 33 30 93.5 62 60 180.0 (11], [12]], [[32], [33) 194.00 ... 0 38-83 (D2-B5) MS3/c01n05_gigue.mscx xml Instrument 1 59-83 (B3-B5) Instrument 1 38-64 (D2-E4) <NA> <NA>
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
parnasse_03 {1: '4/4'} {1: 2} 59 58 235.0 59 58 235.0 () 655.00 ... 0 38-86 (D2-D6) MS3/parnasse_03.mscx <NA> Piano 61-86 (C#4-D6) Piano 57-84 (A3-C6) Piano 38-67 (D2-G4)
parnasse_04 {1: '3/8'} {1: 2} 57 57 84.5 57 57 84.5 () 196.62 ... 0 40-83 (E2-B5) MS3/parnasse_04.mscx <NA> Piano 62-83 (D4-B5) Piano 57-83 (A3-B5) Piano 40-69 (E2-A4)
parnasse_05 {1: '4/4'} {1: 2} 27 26 107.5 27 26 107.5 () 276.00 ... 0 36-83 (C2-B5) MS3/parnasse_05.mscx <NA> Piano 59-83 (B3-B5) Piano 55-83 (G3-B5) Piano 36-66 (C2-F#4)
parnasse_06 {1: '3/8'} {1: 2} 32 32 48.0 32 32 48.0 () 121.63 ... 0 38-83 (D2-B5) MS3/parnasse_06.mscx <NA> Piano 65-83 (E#4-B5) Piano 64-83 (E4-B5) Piano 38-66 (D2-F#4)
parnasse_07 {1: '4/4'} {1: 2} 53 53 212.0 53 53 212.0 () 514.00 ... 0 38-86 (D2-D6) MS3/parnasse_07.mscx <NA> Piano 61-84 (C#4-C6) Piano 59-86 (B3-D6) Piano 38-69 (D2-A4)

84 rows × 53 columns

mean_composition_years = utils.corpus_mean_composition_years(all_metadata)
chronological_order = mean_composition_years.index.to_list()
corpus_colors = dict(zip(chronological_order, utils.CORPUS_COLOR_SCALE))
corpus_names = {
    corp: utils.get_corpus_display_name(corp) for corp in chronological_order
}
chronological_corpus_names = list(corpus_names.values())
corpus_name_colors = {
    corpus_names[corp]: color for corp, color in corpus_colors.items()
}
mean_composition_years
corpus
couperin_concerts    1723.380952
Name: mean_composition_year, dtype: float64

Composition dates#

This section relies on the dataset’s metadata.

valid_composed_start = pd.to_numeric(all_metadata.composed_start, errors="coerce")
valid_composed_end = pd.to_numeric(all_metadata.composed_end, errors="coerce")
print(
    f"Composition dates range from {int(valid_composed_start.min())} {valid_composed_start.idxmin()} "
    f"to {int(valid_composed_end.max())} {valid_composed_end.idxmax()}."
)
Composition dates range from 1722 ('couperin_concerts', 'c01n01_prelude') to 1724 ('couperin_concerts', 'c05n01_prelude').

Mean composition years per corpus#

def make_summary(metadata_df):
    piece_is_annotated = metadata_df.label_count > 0
    return metadata_df[piece_is_annotated].copy()
Hide source
summary = make_summary(all_metadata)
bar_data = pd.concat(
    [
        mean_composition_years.rename("year"),
        summary.groupby(level="corpus").size().rename("pieces"),
    ],
    axis=1,
).reset_index()

N = len(summary)
fig = px.bar(
    bar_data,
    x="year",
    y="pieces",
    color="corpus",
    color_discrete_map=corpus_colors,
    title=f"Temporal coverage of the {N} annotated pieces in the Distant Listening Corpus",
)
fig.update_traces(width=5)
fig.update_layout(**utils.STD_LAYOUT)
fig.update_traces(width=5)
save_figure_as(fig, "pieces_timeline_bars")
fig.show()
summary
TimeSig KeySig last_mc last_mn length_qb last_mc_unfolded last_mn_unfolded length_qb_unfolded volta_mcs all_notes_qb ... has_drumset ambitus path originalFormat staff_1_instrument staff_1_ambitus staff_2_instrument staff_2_ambitus staff_3_instrument staff_3_ambitus
corpus piece
couperin_concerts c01n01_prelude {1: '4/4'} {1: 1} 25 23 98.0 25 23 98.0 () 219.00 ... 0 36-84 (C2-C6) MS3/c01n01_prelude.mscx xml Instrument 1 59-84 (B3-C6) Instrument 1 36-57 (C2-A3) <NA> <NA>
c01n02_allemande {1: '4/4'} {1: 1} 20 18 72.0 40 36 144.0 () 168.50 ... 0 31-83 (G1-B5) MS3/c01n02_allemande.mscx xml Instrument 1 59-83 (B3-B5) Instrument 1 31-66 (G1-F#4) <NA> <NA>
c01n03_sarabande {1: '3/4'} {1: -2} 30 28 90.0 56 56 168.0 (8], [9]], [[29], [30) 237.75 ... 0 31-81 (G1-A5) MS3/c01n03_sarabande.mscx xml Instrument 1 58-81 (Bb3-A5) Instrument 1 31-67 (G1-G4) <NA> <NA>
c01n04_gavotte {1: '2/2'} {1: -2} 18 14 60.0 32 28 112.0 (5], [6]], [[17], [18) 135.00 ... 0 38-79 (D2-G5) MS3/c01n04_gavotte.mscx xml Instrument 1 58-79 (Bb3-G5) Instrument 1 38-57 (D2-A3) <NA> <NA>
c01n05_gigue {1: '6/8'} {1: 1} 33 30 93.5 62 60 180.0 (11], [12]], [[32], [33) 194.00 ... 0 38-83 (D2-B5) MS3/c01n05_gigue.mscx xml Instrument 1 59-83 (B3-B5) Instrument 1 38-64 (D2-E4) <NA> <NA>
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
parnasse_03 {1: '4/4'} {1: 2} 59 58 235.0 59 58 235.0 () 655.00 ... 0 38-86 (D2-D6) MS3/parnasse_03.mscx <NA> Piano 61-86 (C#4-D6) Piano 57-84 (A3-C6) Piano 38-67 (D2-G4)
parnasse_04 {1: '3/8'} {1: 2} 57 57 84.5 57 57 84.5 () 196.62 ... 0 40-83 (E2-B5) MS3/parnasse_04.mscx <NA> Piano 62-83 (D4-B5) Piano 57-83 (A3-B5) Piano 40-69 (E2-A4)
parnasse_05 {1: '4/4'} {1: 2} 27 26 107.5 27 26 107.5 () 276.00 ... 0 36-83 (C2-B5) MS3/parnasse_05.mscx <NA> Piano 59-83 (B3-B5) Piano 55-83 (G3-B5) Piano 36-66 (C2-F#4)
parnasse_06 {1: '3/8'} {1: 2} 32 32 48.0 32 32 48.0 () 121.63 ... 0 38-83 (D2-B5) MS3/parnasse_06.mscx <NA> Piano 65-83 (E#4-B5) Piano 64-83 (E4-B5) Piano 38-66 (D2-F#4)
parnasse_07 {1: '4/4'} {1: 2} 53 53 212.0 53 53 212.0 () 514.00 ... 0 38-86 (D2-D6) MS3/parnasse_07.mscx <NA> Piano 61-84 (C#4-C6) Piano 59-86 (B3-D6) Piano 38-69 (D2-A4)

84 rows × 53 columns

Composition years histogram#

Hide source
hist_data = summary.reset_index()
hist_data.corpus = hist_data.corpus.map(corpus_names)
fig = px.histogram(
    hist_data,
    x="composed_end",
    color="corpus",
    labels=dict(
        composed_end="decade",
        count="pieces",
    ),
    color_discrete_map=corpus_name_colors,
    title=f"Temporal coverage of the {N} annotated pieces in the Distant Listening Corpus",
)
fig.update_traces(xbins=dict(size=10))
fig.update_layout(**utils.STD_LAYOUT)
fig.update_legends(font=dict(size=16))
save_figure_as(fig, "pieces_timeline_histogram", height=1250)
fig.show()

Dimensions#

Overview#

def make_overview_table(groupby, group_name="pieces"):
    n_groups = groupby.size().rename(group_name)
    absolute_numbers = dict(
        measures=groupby.last_mn.sum(),
        length=groupby.length_qb.sum(),
        notes=groupby.n_onsets.sum(),
        labels=groupby.label_count.sum(),
    )
    absolute = pd.DataFrame.from_dict(absolute_numbers)
    absolute = pd.concat([n_groups, absolute], axis=1)
    sum_row = pd.DataFrame(absolute.sum(), columns=["sum"]).T
    absolute = pd.concat([absolute, sum_row])
    return absolute


absolute = make_overview_table(summary.groupby("workTitle"))
# print(absolute.astype(int).to_markdown())
absolute.astype(int)
pieces measures length notes labels
Concert Royal no. 1 6 137 491 1859 480
Concert Royal no. 2 5 248 711 2500 597
Concert Royal no. 3 8 300 865 3565 805
Concert Royal no. 4 7 242 780 2743 652
Le Parnasse, ou L'Apothéose de Corelli 7 306 941 4825 957
Nouveau Concert, ou Les Goûts-Réunis no. 10 4 124 431 1596 364
Nouveau Concert, ou Les Goûts-Réunis no. 11 8 254 964 2890 855
Nouveau Concert, ou Les Goûts-Réunis no. 14 4 122 408 1645 466
Nouveau Concert, ou Les Goûts-Réunis no. 5 5 205 617 2014 529
Nouveau Concert, ou Les Goûts-Réunis no. 6 5 164 566 2133 590
Nouveau Concert, ou Les Goûts-Réunis no. 7 6 171 637 2364 602
Nouveau Concert, ou Les Goûts-Réunis no. 8 11 420 1620 3542 939
Nouveau Concert, ou Les Goûts-Réunis no. 9 8 252 800 3211 919
sum 84 2945 9834 34887 8755
def summarize_dataset(D):
    all_metadata = D.get_metadata()
    summary = make_summary(all_metadata)
    return make_overview_table(summary.groupby(level=0))


corpus_summary = summarize_dataset(D)
print(corpus_summary.astype(int).to_markdown())
|                   |   pieces |   measures |   length |   notes |   labels |
|:------------------|---------:|-----------:|---------:|--------:|---------:|
| couperin_concerts |       84 |       2945 |     9834 |   34887 |     8755 |
| sum               |       84 |       2945 |     9834 |   34887 |     8755 |

Measures#

all_measures = D.get_feature("measures")
print(
    f"{len(all_measures.index)} measures over {len(all_measures.groupby(level=[0,1]))} files."
)
all_measures.head()
3394 measures over 91 files.
mc mn quarterbeats duration_qb keysig timesig act_dur mc_offset numbering_offset dont_count barline breaks repeats next volta markers jump_bwd jump_fwd play_until
corpus piece i
couperin_concerts c01n01_prelude 0 1 0 0 2.0 1 4/4 1/2 1/2 <NA> 1 <NA> <NA> firstMeasure (2,) <NA> <NA> <NA> <NA> <NA>
1 2 1 2 4.0 1 4/4 1 0 <NA> <NA> <NA> <NA> <NA> (3,) <NA> <NA> <NA> <NA> <NA>
2 3 2 6 4.0 1 4/4 1 0 <NA> <NA> <NA> <NA> <NA> (4,) <NA> <NA> <NA> <NA> <NA>
3 4 3 10 4.0 1 4/4 1 0 <NA> <NA> <NA> line <NA> (5,) <NA> <NA> <NA> <NA> <NA>
4 5 4 14 4.0 1 4/4 1 0 <NA> <NA> <NA> <NA> <NA> (6,) <NA> <NA> <NA> <NA> <NA>
all_measures.get_default_analysis().plot_grouped()

Harmony labels#

All symbols, independent of the local key (the mode of which changes their semantics).

try:
    all_annotations = D.get_feature("harmonylabels").df
except Exception:
    all_annotations = pd.DataFrame()
n_annotations = len(all_annotations.index)
includes_annotations = n_annotations > 0
if includes_annotations:
    display(all_annotations.head())
    print(f"Concatenated annotation tables contains {all_annotations.shape[0]} rows.")
    no_chord = all_annotations.root.isna()
    if no_chord.sum() > 0:
        print(
            f"{no_chord.sum()} of them are not chords. Their values are:"
            f" {all_annotations.label[no_chord].value_counts(dropna=False).to_dict()}"
        )
    all_chords = all_annotations[~no_chord].copy()
    print(
        f"Dataset contains {all_chords.shape[0]} tokens and {len(all_chords.chord.unique())} types over "
        f"{len(all_chords.groupby(level=[0,1]))} documents."
    )
    all_annotations["corpus_name"] = all_annotations.index.get_level_values(0).map(
        utils.get_corpus_display_name
    )
    all_chords["corpus_name"] = all_chords.index.get_level_values(0).map(
        utils.get_corpus_display_name
    )
else:
    print("Dataset contains no annotations.")
mc mn quarterbeats duration_qb mc_onset mn_onset timesig staff voice volta ... numeral_or_applied_to_numeral intervals_over_bass intervals_over_root scale_degrees scale_degrees_and_mode scale_degrees_major scale_degrees_minor globalkey localkey chord
corpus piece i
couperin_concerts c01n01_prelude 0 1 0 0 2.00 0 1/2 4/4 1 1 <NA> ... I (M3, P5) (M3, P5) (1, 3, 5) (1, 3, 5), major (1, 3, 5) (1, #3, 5) G I I
1 2 1 2 2.00 0 0 4/4 1 1 <NA> ... V (M3, P5) (M3, P5) (5, 7, 2) (5, 7, 2), major (5, 7, 2) (5, #7, 2) G I V
2 2 1 4 0.50 1/2 1/2 4/4 1 1 <NA> ... I (m3, m6) (M3, P5) (3, 5, 1) (3, 5, 1), major (3, 5, 1) (#3, 5, 1) G I I6
3 2 1 9/2 0.50 5/8 5/8 4/4 1 1 <NA> ... I (M3, P5) (M3, P5) (1, 3, 5) (1, 3, 5), major (1, 3, 5) (1, #3, 5) G I I
4 2 1 5 0.75 3/4 3/4 4/4 1 1 <NA> ... V (P4, P5) (P4, P5) (5, 1, 2) (5, 1, 2), major (5, 1, 2) (5, 1, 2) G I V(4)

5 rows × 51 columns

Concatenated annotation tables contains 8376 rows.
Dataset contains 8376 tokens and 241 types over 84 documents.